US20080228843A1

US20080228843A1 - Filesystem having a filename cache

Info

Publication number: US20080228843A1
Application number: US11/717,583
Authority: US
Inventors: Dan Dodge; Peter Van Der Veen
Original assignee: Individual
Current assignee: 8758271 Canada Inc; Malikie Innovations Ltd
Priority date: 2006-08-25
Filing date: 2007-03-13
Publication date: 2008-09-18
Also published as: US20110078219A1; US8122178B2; US20110246477A1; US7987190B2; US7908276B2

Abstract

A system comprising a processor, a data storage device that is accessible by the processor, and filesystem software that is executable by the processor to organize files on the data storage device are provided. The filesystem software is executable to maintain a filename cache having variable size filename entries. The filename entries may include a filename header section and a name section storing an entire filename of a corresponding file.

Description

BACKGROUND OF THE INVENTION

1. Technical Field
This invention is generally directed to a filesystem for use in a computer, embedded controller, processing system, or the like. More particularly, this invention is directed to a filesystem having a filename cache.
2. Related Art
Computers, embedded controllers, and other microprocessor based systems are typically constructed from a variety of different hardware components. The hardware components may include a processor, I/O devices, human interface devices, and the like. Additionally, such systems use memory storage units to maintain the data used in the system. The memory storage units may take on a variety of different forms including, but not limited to, hard disk drives, floppy disk drives, random access memory, flash memory, and the like.
High-level application programs that are executed in such systems must often interact seamlessly with these hardware components, including the memory storage units. To this end, many systems run an operating system that acts as an interface between the application programs and the system hardware. Filesystem software may be included as part of the operating system or it may be provided as an ancillary software component that interacts with the operating system. In either instance, the filesystem software organizes the data within the memory storage units for ready access by the processor and the high-level application programs that the processor executes.
The filesystem software may employ a file/directory layer that organizes the contents of files and directories into equal-sized logical blocks of contiguous data. Each logical block has an association with one or more corresponding physical blocks on the storage device where the data is actually retained. The file/directory layer executes updates to the filesystem by identifying every block that needs to be updated in response to a request and rewriting the entire contents of each such block. The file/directory layer reads the contents of files and directories by reading the entire contents of every block that holds a portion of the region of data to be read.
The filesystem also may include a storage layer that maps the virtual addresses of filesystem contents to physical blocks of data on the data storage device. The storage layer may execute logical block read requests from the file/directory layer by determining the correct physical block(s) associated with the request and reading its contents from the data storage device. Similarly, the storage layer may execute right requests by either updating contents of an existing physical block(s), or by allocating an unused physical block from the data storage device to the logical block and then updating the contents of the physical block.
The filesystem also may be responsible for locating a file in data storage using, for example, a filename associated with the file. Systems such as UNIX employ a filename cache in which a fixed number of bytes are allocated to each filename for this purpose. This system may be wasteful of memory space when the filenames are short and may result in ambiguous hits for long filenames that have been truncated to fit into the fixed byte allocation. Accordingly, there is a need for an improved filename caching system.

SUMMARY

A system that comprises a processor, a data storage device that is accessible by the processor, and filesystem software that is executable by the processor to organize files on the data storage device are provided. The filesystem software may be executable to maintain a filename cache having variable size filename entries. The filename entries may include a filename header section and a name section storing an entire filename of a corresponding file. In one example, the filename cash is implemented in a ring-type buffer or the like.
The filename header section may include a variety of information. For example, the filename header section may include information corresponding to a length of the filename in the name section, information uniquely identifying the file associated with the filename in the name section, information for indexing into a filename hash table, information indicative of whether the filename entry corresponds to a first entry in the filename hash table for a particular namesum, and/or information indicative of whether the filename in the respective name section is not locatable on the system. Still further, the filename header section may include an index to the next name in the filename cache buffer having the same namesum value to thereby facilitate searches through the filename cache.
Other systems, methods, features and advantages of the invention will be, or will become, apparent to one with skill in the art upon examination of the following figures and detailed description. It is intended that all such additional systems, methods, features and advantages be included within this description, be within the scope of the invention, and be protected by the following claims.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention can be better understood with reference to the following drawings and description. The components in the figures are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the invention. Moreover, in the figures, like referenced numerals designate corresponding parts throughout the different views.

FIG. 1 is a block diagram of a processing system that may implement a filesystem having a filename cache.

FIG. 2 illustrates one manner in which the filesystem software may implement a filename cache.

FIG. 3 illustrates one manner of arranging information in the filename header section of each filename entry shown in FIG. 2.

FIG. 4 is a flow chart showing a number of interrelated operations that may be used by the filesystem software to respond to a filename access request.

FIG. 5 is a flow chart showing a number of interrelated operations that may be executed by the filesystem software in responding to a filename cache update request.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

FIG. 1 illustrates the components that may be employed in an exemplary processing system 100. As shown, the exemplary system 100 includes a processor 105, read only memory 110, and data storage 115. Processing system 100 also may include random access memory 120, an I/O interface 125, and a user interface 130. The specific components that are used in processing system 100 may be tailored to the particular function(s) that are to be executed by the processing system 100. Accordingly, the presence or absence of a component may be specific to the design criterion imposed on the processing system 100.
Data storage 115 may include operating system code 135 that controls the interaction between high-level application programs executed by the processor 105 and the various hardware components, including memory 110 and 120, the data storage 115, and the interface devices 125 and 130. The operating system code 135 may include filesystem software for organizing files stored on the data storage 115. Alternatively, the filesystem software may be provided as a separate software component that merely interacts with the operating system code 135. In the latter case, the code corresponding to the filesystem software may be stored in read only memory 110, data storage 115 or the like. When processing system 100 is networked with other computers and/or storage devices through I/O interface 125, the filesystem software may be stored remotely and downloaded to the processing system 100 as needed. FIG. 1, however, illustrates storage of the filesystem software 140 in data storage 115.
The data storage 115 may take on any number of different forms. For example, the data storage 115 may take the form of a hard disk drive, floppy disk drive, etc. It also may be in the form of a non-rotating media device, such as non-volatile memory implemented in an integrated circuit format (e.g., flash memory, and the like). Still further, data storage 115 need not be limited to a single memory structure. Rather, the data storage 115 may include a number of separate storage devices of the same type (e.g., all flash memory) and/or separate storage devices of different types (e.g., one or more flash memory units and one or more hard disk drives).
The files stored in the data storage 115 include data that is interpreted in accordance with a predetermined format used by an application program or by the operating system code 135. For example, the data stored within a file may constitute the software code of an executable program, the ASCII text of a database record, audio media files, video media files, or the like.
Storage 115 also includes a metadata file 145 that stores metadata information for the files organized in the file data area 150. In other system designs, metadata file 145 may be stored in RAM 120. Access to the metadata information for a file through use of a filename is streamlined using a hash table 155 and filename cache 160. In the exemplary system shown in FIG. 1, both the hash table 155 and the filename cache 160 are stored in RAM 120.
FIG. 2 illustrates one manner in which the filesystem software 140 may implement filename cache 160 and hash table 155. In this example, filename cache 160 may be in the form of a fixed sized buffer having filename entries 205 of varying size. Each filename entry 205 is comprised of a filename header section 210 and a name section 215. Except in the case of dummy filename entries, the name section 215 of each filename entry 205 includes the entire filename for a respective file and is variable in size while the filename header section 210 may have a fixed size.
FIG. 3 illustrates one manner of arranging information in the filename header section 210. As shown, the filename header 210 includes a “next cache location” field 305, a unique file identifier field 310, a sequence number field 315, a name length field 320, a back index field 325, a missing field 330, an alias field 335, and an IsBack field 340. The “next cache location” field 305 serves as an index/pointer to the next filename entry in the cache 160 that has a filename with the same namesum value. This field is used to search the filename cache for a particular filename while avoiding unnecessary checks of intermediate filename entries that do not have the same namesum value. The unique file identifier field 310 is used to uniquely identify a given file in the filesystem. Here, the unique file identifier field 310 corresponds to the file having the filename in the name section 215 of the filename entry 205 and may be used to differentiate between different files having the same filename but which are stored in different storage locations. The sequence field 315 includes a sequence number that is unique to the filesystem. The name length field 320 of header section 210 provides information on the length of the name section 215, such as the number of bytes contained in the name section. The back field 325 provides an index into the hash table for the particular filename entry 205. The missing field 330 is used to identify whether the filename in name section 215 is located on the system. This allows the filesystem software 140 to quickly identify filenames that already have been searched and not located on the system thereby allowing the filesystem software to forgo unnecessary searching for the missing filename. The alias field 335 is used to identify the contents of the name section 215 as an alias name for the file. The IsBack field 340 is used to indicate whether the filename entry 205 corresponds to the first entry in hash table 155 for a particular namesum.
FIG. 4 is a flow chart showing a number of interrelated operations that may be used by the filesystem software 140 to respond to a filename access request. In this example, a filename access request is made at block 405. The request may include the filename and an identification of the parent directory/directories. The filesystem software 140 proceeds to calculate a namesum using the requested filename at block 410. A hash function is executed on the calculated namesum at block 415 to generate an index into hash table 155. The hash index is used at block 420 to obtain an initial pointer into the filename cache 160. If the initial pointer is invalid, then the filesystem software 140 assumes that the filename has not been cached at block 425 and an alternate filename search outside of the filename cache 160 is executed at block 430. Otherwise, the filesystem software 140 assumes that the filename is in the filename cache 160 and begins searching the filename entries 205 at block 435.
One manner of searching through the filename entries 205 is represented in FIG. 2. To this end, the index into hash table 155 obtained at block 415 of FIG. 4 is shown at arrow 220 and points to a hash table location 225 corresponding to a namesum=y (assuming that the namesum calculated at block 410 of FIG. 4 is also y.) The data in hash table location 225 serves as an initial pointer that directs the filesystem software 140 to the first entry in the filename cache 160 that has a name section 215 having the same namesum (y) as the filename requested at block 405. Here, the first entry is designated at 230 and includes filename header section 235 and name section 240. The filesystem software 140 compares the complete filename received at block 405 of FIG. 4 to the complete filename stored in name section 240 to determine whether there is a match. If there is no match, the filesystem software 140 checks the next field 305 of filename header section 235 to identify the next entry in the filename cache 160 storing a filename having a namesum=y. In FIG. 2, the next filename entry with namesum=y is filename entry 245 having filename header section 250 and name section 255. Again, the complete filename stored in name section 240 is compared to the complete name received at block 405 to determine whether there is a match. Assuming that there is a match, the unique file identifier 310 of filename header section 250 is obtained at block 440 of FIG. 4 and used at block 445 to determine whether the file referenced by the unique file identifier is a child of the parent directory provided during the filename access request of block 405. If it is, the filename has been found as indicated at block 450. If it is not, the filesystem software checks at block 455 to determine whether all appropriate filename entries having a namesum=y have been searched. If less than all appropriate filename entries have been searched, the search continues at block 460, where the next field 305 of the current filename entry is used to identify the next filename entry that is to be compared. In FIG. 2, the next field 305 of filename header section 250 points to filename entry 260 having filename header section 265 and name section 270. If all appropriate filename entries have been searched as determined at block 455, the alternate filename search of block 430 may be executed.
FIG. 5 is a flow chart showing a number of interrelated operations that may be executed by the filesystem software 140 in responding to a filename cache update request. As shown, a filename cache update request is received at block 505 and, in the illustrated example, may correspond to one of three different request types: filename found update, filename not on system update, and filename deleted update. In a “filename found update,” the filesystem software 140 is directed to store a filename entry for a corresponding filename in the filename cache 160. To this end, the information provided to the filesystem software 140 in the request is used to generate a filename header section and name section for the filename entry at block 510. Since the filename cache 160 is in the form of a fixed buffer, it employs a put pointer 275 (see FIG. 2) that rotates through the cache buffer to indicate where the next filename entry is to be stored. Once the put pointer extends beyond the end of the filename cache 160, it is rotated back to a beginning portion of the filename cache 160. As such, filename entries are added to the filename cache 160 in a ring-like manner where newer filename entries in the cache are written over older filename entries.
The filename header section and name section for the filename entry generated at block 510 are stored as a filename entry at the put pointer location in the operation shown at block 515. This operation may overwrite one or more existing filename entries of the filename cache 160 thereby leaving unused space between the end of the new filename entry and the beginning of the next complete filename entry in the filename cache 160. To reduce problems that might otherwise be associated with this unused space, the unused space may be allocated to a dummy file name entry. To this end, the filesystem software checks at block 520 to determine whether a dummy entry is needed. If needed, the dummy entry is generated and stored after the current filename entry in the operation at block 525. If no dummy entry is needed, the put pointer 275 is updated to point to the end of the newly added filename entry in the operation at block 530. Alternatively, if a dummy entry is used at block 525, the put pointer 275 is updated to point to the end of the newly generated dummy entry. In the operation at block 535, the hash table 155 is updated to reflect changes to the filename cache buffer.
In a “filename not on system” update, the filesystem software 140 is directed to store a filename entry for a corresponding filename in the filename cache 160 and indicate that it is not on the system. To this end, the filename provided at block 505 is used to generate a filename header and name section for the absent filename in the operation shown at block 540. The missing field 330 of the generated filename header will be set to indicate that the filename is not on the system. This allows the filesystem software 140 to forgo searching the filesystem for a filename that has already been searched and not found on the system. Once the operation at block 540 is complete, the filesystem software 140 may proceed to execute the operations shown in blocks 515 through 535.
In a “filename deleted” update, the filesystem software 140 is directed to remove a filename entry for a corresponding filename from the filename cache 160. As shown, this is accomplished at block 545. In the operations at block 545, one or more bits in the filename header section of the filename entry that is to be removed are changed to indicate removal of the filename. For example, the unique file identifier field 310 may be filled with zeros or another value that is not associated with an actual file and known to be indicative of a removed file. Once the corresponding filename header section has been changed, the hash table is updated in the operations at block 535 to reflect changes to the filename cache 160.
While various embodiments of the invention have been described, it will be apparent to those of ordinary skill in the art that many more embodiments and implementations are possible within the scope of the invention. Accordingly, the invention is not to be restricted except in light of the attached claims and their equivalents.

Claims

1. A system comprising:

a processor;

a data storage device accessible by the processor; and

filesystem software executable by the processor to organize files on the data storage device, where the filesystem software is executable to maintain a filename cache having variable size filename entries, and where the variable size filename entries include a filename header section and a name section storing an entire filename of a corresponding file.

2. The system of claim 1, where the filename cache is stored in a ring-type buffer.

3. The system of claim 1, where the filename header section comprises information corresponding to a length of the filename in the name section.

4. The system of claim 1, where the filename header section comprises information uniquely identifying a file associated with the filename in the name section.

5. The system of claim 1, where the filename header section comprises information for indexing into a filename hash table.

6. The system of claim 5, where the filename header section comprises information indicative of whether the variable size filename entry corresponds to a first entry in the filename hash table for a particular namesum value.

7. The system of claim 1, where the filename header section comprises information indicative of whether the filename in the respective name section is not locatable on the system.

8. The system of claim 1, where the filename header section comprises an index to a next name in the filename cache buffer having a same namesum value.

9. A system comprising:

a filename cache; and

filesystem software organizing the filename cache into variable size filename entries, where the variable size filename entries include a filename header section and a name section storing an entire filename of a corresponding file.

10. The system of claim 9, where the filename cache comprises a ring-type buffer.

11. The system of claim 10, where the filename cache comprises dummy filename entries to compensate for wrap-around writing of filename entries to the ring-type buffer.

12. The system of claim 9, where the filesystem software calculates a namesum using a filename of a requested file.

13. The system of claim 12, where the filesystem software uses the namesum to generate an index into a hash table.

14. The system of claim 13, where the filesystem software uses the index into the hash table to obtain an initial pointer from the hash table that directs the filesystem software to a first variable size filename entry in the filename cache having the calculated namesum.

15. The system of claim 14, where the filesystem software compares the requested filename to the filename stored in the name section of the first variable size filename entry for a match.

16. The system of claim 15, where the filename header section of the first variable size filename entry comprises information corresponding to a parent directory of the corresponding file.

17. The system of claim 16, where the filesystem software compares parent directory information of the requested file to the parent directory information of the filename header section of the first variable size filename entry to determine whether there is a match between the requested file and the file corresponding to the first variable size filename entry.

18. The system of claim 9, where the filename header section comprises information corresponding to a length of the filename in the name section.

19. The system of claim 9, where the filename header section comprises information uniquely identifying a file associated with the variable size filename entry.

20. The system of claim 9, where the filename header section comprises information for indexing into a filename hash table.

21. The system of claim 20, where the filename header section comprises information indicative of whether the variable size filename entry corresponds to a first entry in the filename hash table for a particular namesum.

22. The system of claim 9, where the filename header section comprises information indicative of whether the filename in the respective name section is not locatable on the system.

23. The system of claim 9, where the filename header section comprises an index to a next name in the filename cache buffer having a same namesum value.

24. The system of claim 9, where the filesystem software generates a variable size filename entry in the filename cache for a requested file when the filename cache does not include an entry for the requested file and the requested file is located by the filesystem software.

25. The system of claim 9, where the filesystem software updates one or more bits of a variable size filename entry to indicate that a filename stored in the name section of the variable size filename entry cannot be located by the filesystem software.

26. The system of claim 25, where the filesystem software checks the one or more bits to discontinue further searching for a filename that has been previously searched and not found by the filesystem software.

27. The system of claim 25, where the one or more bits are stored in the header section of the variable size filename entry.

28. The system of claim 9, where the filesystem software removes a variable size filename entry for a corresponding file when the corresponding file is deleted from the system.

29. A method for operating a filesystem comprising:

allocating storage space to a filename cache; and

organizing the filename cache into variable size filename entries, where the variable size filename entries include a filename header section and a name section storing an entire filename of a corresponding file.

30. The method of claim 29, where the allocated storage space for the filename cache comprises a ring-type buffer.

31. The method of claim 30, further comprising adding dummy filename entries to compensate for wrap-around writing of variable size filename entries to the ring-type buffer.

32. The method of claim 29, further comprising calculating a namesum using a filename of a requested file.

33. The method of claim 32, further comprising using the calculated namesum to generate an index into a hash table.

34. The method of claim 33, further comprising:

using the index into the hash table to obtain an initial pointer from the hash table; and

using the initial pointer to access a first variable size filename entry in the filename cache having the calculated namesum.

35. The method of claim 34, further comprising comparing the requested filename to the filename stored in the name section of the first variable size filename entry for a match.

36. The method of claim 35, where the filename header section of the first variable size filename entry comprises information corresponding to a parent directory of the corresponding file.

37. The method of claim 36, further comprising comparing parent directory information of the requested file to the parent directory information of the filename header section of the first variable size filename entry to determine whether there is a match between the requested file and the file corresponding to the first variable size filename entry.

38. The method of claim 29, further comprising generating a variable size filename entry in the filename cache for a requested file when the filename cache does not include an entry for the requested file and the requested file is located by the filesystem.

39. The method of claim 29, further comprising updating one or more bits of a variable size filename entry to indicate that a filename stored in the name section of the variable size filename entry cannot be located by the filesystem.

40. The method of claim 39, further comprising checking the one or more bits to discontinue further searching for a filename that has been previously searched and not found by the filesystem.

41. The method of claim 40, where the one or more bits are stored in the header section of the variable size filename entry.

42. The method of claim 29, further comprising removing a variable size filename entry when the corresponding file is deleted from the filesystem.

43. A storage medium comprising code adapted to execute a method for operating a filesystem comprising:

allocating storage space to a filename cache; and

44. The storage medium of claim 43, where the storage space allocated for the filename cache by the method comprises a ring-type buffer.

45. The storage medium of claim 44, where the method further comprises adding dummy filename entries to compensate for wrap-around writing of variable size filename entries to the ring-type buffer.

46. The storage medium of claim 43, where the method further comprises calculating a namesum using a filename of a requested file.

47. The storage medium of claim 46, where the method further comprises using the calculated namesum to generate an index into a hash table.

48. The storage medium of claim 47, where the method further comprises:

49. The storage medium of claim 48, where the method further comprises comparing the requested filename to the filename stored in the name section of the first variable size filename entry for a match.

50. The storage medium of claim 49, where the filename header section of the first variable size filename entry employed by the method comprises information corresponding to a parent directory of the corresponding file.

51. The storage medium of claim 50, where the method further comprises comparing parent directory information of the requested file to the parent directory information of the filename header section of the first variable size filename entry to determine whether there is a match between the requested file and the file corresponding to the first variable size filename entry.